The present invention relates generally to scintillators. More specifically, the present invention provides lanthanum bromide (LaBr3) scintillators for use with medical imaging scanner systems, such as time-of-flight PET scanner systems.
One form of medical imaging is called positron emission tomography and is better known by its acronym PET. In PET (or PET scan), the patient is injected with or otherwise administered a molecule labeled with a positron-emitting radioactive element. In some applications the radiotracer is distributed through the body, and concentrated in (or excluded from) target tissues of interest. The radioactive material decays by emission of a positron, or antiparticle of the negatively-charged electron. The positron is slowed down within a short distance from the emission point and forms a short-lived “atom” consisting of the positron and an electron from a nearby atom. The “atom,” referred to as positronium, decays by the annihilation of its constituents. This annihilation produces two essentially back to back 511 keV gamma rays which are emitted in substantially opposite directions. When both of the gamma rays are detected by detectors surrounding the body, it can be assumed with high probability that the emission point was somewhere along a line joining the two detectors. Without additional information, the probability that the radiotracer was located on any one point in the body that the detection line intersects is equal for all points in the line, and hence, for all points in the body being scanned.
A variety of algorithms have been developed that make it possible to form an image from a collection of such lines. The quality of the image improves, in general, as the number of lines increases. Similarly, as the signal-to-noise of the image depends on the square root of the number of lines, each line representing one annihilation event, more lines offer an improved signal to noise ratio. Nevertheless, a common aspect of all image formation or reconstruction algorithms is that the noise increases in the process of deciding where along the detected line the annihilation event is likely to have occurred. In one aspect, this effect can be thought of in terms of energy and work: The detected lines represent the energy in the image and a large part of that energy is used up as work in localizing the annihilation along the particular line, instead of contributing to image quality. If the body cross-section is 30 cm and the desired localization accuracy of an annihilation event is 5 mm, localization requires reducing the uncertainty of its location by a factor of 60.
The annihilation gamma-rays travel at a speed of about 30 cm/nanosecond (1 foot/ns). The timing accuracy of detectors currently used commercially in PET cameras is a few nanoseconds (ns). Timing resolution is typically applied to two aspects of PET: One use is to reduce accidentals (the overlapped detection of two unrelated gamma-rays). The other use is in timing signals for localization purposes. While any improvement in time resolution aids in accidentals reduction, until the time resolution drops substantially below about 1 ns, greater time resolution will not help in localization of an annihilation event within a target of about 30 cm—the more common presentation in human-PET scanning. That is, if a body cross-section is about 30 cm, that very fact localizes the event without any recourse to time resolution. With that limitation, an improvement in localization from 4 nanosecondss (4 feet) to 1 nanosecond (1 foot) offers no improvement to image quality. On the other hand, a time signal improvement from 1 nanosecond to 500 picoseconds (0.5 nanoseconds) reduces the uncertainty of event location by a factor of 2. To appreciate the value of this particular improvement, it is to be noted that the factor of 2 increase in time resolution accuracy results in a corresponding increase in signal-to-noise ratio. This results in the equivalent of a factor of 4 increase in detected annihilation events. Placed in a different context, under the same circumstances, an image (actually a data set) that may take 15 minutes to obtain with a 1 nanoseconds time resolution, is obtained in under about 4 minutes when the time resolution is 500 picoseconds.
FIG. 1 illustrates the principle on which the location along the detected line is used to improve image formation.
It is generally recognized that a fast timing scintillator in PET cameras will enable time-of-flight PET when the timing accuracy is below 1 nanosecond. Hitherto no true time-of-flight PET device has been enabled. Barium fluoride (BaF2), lutetium orthosilicate (LSO) and bismuth germanate (BGO) have been suggested as potentially useful scintillation materials, but none of these materials has the 500 picosecond (ps) or less time resolution needed to achieve a successful device. BaF2 actually provides a ˜250 ps (FWHM) timing resolution, but it has a low emission intensity for the fast component and emits in blue region of the spectrum where special photomultiplier tubes (PMTs) with quartz windows are required for readout. It is noted that a number of plastic scintillators have a time resolution below 500 picoseconds, but due to inadequate stopping power, (attenuation length at 500 keV is typically greater than 10 cm), these scintillators are not suitable for medical uses.
The present invention provides very fast scintillator materials capable of resolving the positron of an annihilation event within a portion of a human body cross-section (less than 400 picoseconds), specifically, the very fast scintillator material comprises LaBr3 doped with various concentrations of cerium. Particular attention is drawn to the unexpected discovery that the quantity of Ce used to dope LaBr3 improves the timing properties (rise time and timing resolution), stopping power and locational capabilities of the scintillator to provide a scintillator capable of use for actual time-of-flight PET.
Attention is drawn to several references in the field, the teachings of which are incorporated herein by reference (as are all references cited herein):
U.S. Pat. No. 6,362,479, “Scintillation detector array for encoding the energy, position, and time coordinates of gamma ray interactions,” discloses a scintillator-encoding scheme that depends on the differential decay time of various scintillators. The use of LSO-LSO crystals with a time resolution of 1.6 nanoseconds is also discussed. A time resolution of 1.6 nanoseconds is equivalent to an approximately 50 cm uncertainty, which is as large as the cross-sectional dimension of the human body, and not useful in the present context.
U.S. Pat. No. 5,453,623, “Positron emission tomography camera with quadrant-sharing photomultipliers and cross-coupled scintillating crystals.” Discloses arrangement of hardware elements in PET camera and use of scintillators. Only specific scintillator disclosed is BGO.
Moses et al., “Prospects for Time-of-Flight PET using LSO Scintillator,” IEEE Trans. Nucl. Sci. 46:474–478 (1999).
U.S. Pat. No. 5,319,203 and U.S. Pat. No. 5,134,293, both entitled “Scintillator material.” Discloses Cerium fluoride and thallium doped Cerium fluoride as “improved” scintillator material.
U.S. Pat. No. 5,039,858, “Divalent fluoride doped cerium fluoride scintillator.” Discloses additional doped cerium fluoride scintillators.
U.S. Pat. No. 4,510,394, “Material for scintillators.” Discloses barium fluoride as scintillator material.
van Loef et al., “High energy resolution scintillator: Ce3+ activated LaBr3”, Appl. Phys. Lett. 79:1573–1575 (2001).
van Loef et al., “Scintillation properties of LaBr3:Ce3+ crystals: fast, efficient and high-energy-resolution scintillators”, Nucl. Instr. Meth. Physics Res. A 486:254–258 (2002). Discloses certain characteristics of cerium doped LaBr3 compositrons including, light yield, and scintillation decay curve. The rise time and time resolution of the compositrons are not disclosed or suggested.
WO 01/60945, “Scintillator crystals, method for making same, use thereof”, Discloses inorganic scintillator material of the general compositron M1-xCexBr3, where M is selected from lanthanides or lanthanide mixtures of the group consisting of La, Gd, and Y. X is the molar rate of substitution of M with cerium, x being present in an amount of not less than 0.01 mol % and strictly less than 100 mol %. The rise time and time resolution of the various compositrons are not disclosed or suggested.